ATTITUDES AND SOCIAL COGNITIONThe Linguistic Relativity of Person Cognition:An English-Chinese ComparisonCurt Hoffman, Ivy Lau, and David Randy JohnsonUniversity of Alberta, Edmonton, CanadaWe performed an experiment to test the possibility that distinct languages (in this case, English andChinese) are capable of exerting language-specific effects on people's impressions of and memory forother individuals. Parallel English- and Chinese-language descriptions were created of two charactersexemplifying personality schemas with economical labels in English but not in Chinese, and twocharacters exemplifying personality schemas with economical labels in Chinese but not in English.Three groups of subjects participated in the experiment: English monolingual^ Chinese-Englishbilinguals who read and responded in English, and Chinese-English bilinguals who read and re-sponded in Chinese. We predicted that subjects processing the target descriptions in English wouldshow greater evidence of schematic thinking in the case of the two characters representing English-labeled schemas, whereas subjects processing the target descriptions in Chinese would show greaterevidence of schematic thinking in the case of the two characters representing Chinese-labeled sche-mas. This prediction was confirmed. Both impressions and memory were affected when the target'spersonality and behavior conformed to a labeled schema in the subject's language of processing. Theresults are discussed in relation to current social psychological conceptions of schematic processingand in relation to the Whorfian hypothesis of linguistic relativity.The idea that the particular language one speaks importantlyaffects the manner in which one perceives and thinks about theworld—the linguistic relativity hypothesis—has a long butsomewhat checkered history within the disciplines of psychol-ogy, anthropology, linguistics, and philosophy. Benjamin LeeWhorf, this century's most influential proponent of the linguis-tic relativity hypothesis, expressed its central proposition as fol-lows:We dissect nature along lines laid down by our native languages.The categories and types that we isolate from the world of phenom-ena we do not find there because they stare every observer in theface; on the contrary, the world is presented in a kaleidoscopic Duxof impressions which has to be organized by our minds—and thismeans largely by the linguistic systems in our minds.. . . We arethus introduced to a new principle of relativity, which holds thatall observers are not led by the same physical evidence to the samepicture of the universe, unless their linguistic backgrounds are sim-ilar, or can in some way be calibrated. (Whorf, 1956, pp. 213-214)Whorf's writings were the primary inspiration for a smallflurry of behavioral research on the language-cognition relationthat arose in the 1950s and continued into the 1960s, most ofwhich was concerned with the impact of language on colormemory and object classification. By the 1970s, the linguisticWe thank Igor Gavanski, Yvonne Ko, Judy Lau, Maria Tchir, andMary Yu for their assistance with the research reported in this article.Correspondence concerning this article should be addressed to CurtHoffman, Department of Psychology, University of Alberta, Edmonton,Alberta, Canada T6G 2E9.relativity hypothesis had largely fallen into disfavor. This wasdue in part to the vague and sometimes extreme form in whichWhorf had stated the hypothesis. It also occurred because an-thropology had begun to accord more importance to linguisticand cultural universals than to linguistic and cultural differ-ences (e.g., Berlin & Kay, 1969), and psychology had begun toquestion (both on empirical and on conceptual grounds) theassumption that categorization is arbitrary and was moving to-ward the view that the perceptual world contains its own built-in category structure (e.g., Rosch, 1974). Recently, however,there have been signs of renewed interest in the linguistic rela-tivity hypothesis. More sophisticated theoretical analyses of thelanguage-cognition relation (e.g., Bloom, 1981), as well asmethodologically superior empirical work (e.g., Lucy &Shweder, 1979), are helping to clarify the ways in which linguis-tic classifications and linguistic structures can and do partiallyshape the cognitive categories and processes by which we cometo know the world.This revival of interest in the linguistic relativity hypothesisis particularly timely in view of the central conceptual role cur-rently accorded to the schema concept in cognitive, social, anddevelopmental psychology. As Bloom (1981) has argued, lan-guage affects cognitive life in two general ways. First, it influ-ences the development of our repertory of cognitive schemas:In addition ... to developing a large number of schemas free ofthe influence of language, which never come to be labeled, and todeveloping a large number of schemas free of the influence of lan-guage, which come in time to be labeled, but whose internal organi-zation remain unaffected by the fact that they receive labels, thechild will construct or reconstruct a very large number of schemasJournal of Personality and Social Psychology, 1986, Vol, 51, No, 6,1097-1105Copyright 1986 by the American Psychological Association, Inc. 0022-3514/86/S00.7510971098C. HOFFMAN, I. LAU, AND D. JOHNSONexpressly to meet the requirements of linguistic labels. (Bloom,1981, p. 66)An example of this last type of schema is that labeled by theword dog; dogs do not constitute an especially coherent or dis-tinctive category, and were it not for the necessity of learningthe correct use of the word dog, it is unlikely that the childwould ever arrive at this particular equivalence class. Other,more abstract examples include the schemas labeled by thewords sister, amount, and too.Second, those schemas that have linguistic labels enjoy a spe-cial status in our mental life. According to Bloom, this is be-cause the use of verbal symbols not only is a prerequisite toovert social communication but also facilitates the covert self-communication in which we engage to structure our ownthought processes.We seem to call specially upon those of our schemas that havenames, via their names, when we want to disengage particular sche-matic perspectives from the collectivity of our interacting associa-tions, ideas, and experiences and make use of those discrete, struc-tured perspectives on reality as stable points of mental orientationto provide direction